1. Technical Field
The following embodiments generally relate to the calibration of multiple cameras and, more particularly, to a method and apparatus for calibrating multiple cameras using mirrors.
2. Description of the Related Art
With the fall in price of cameras and the improvement of computer performance and network performance, multiple cameras, rather than a single camera, have been used to construct a computer vision system.
Multiple cameras may be arranged to suit an actual vision application system, unlike image information from a single camera having a limited field (angle) of view, and a large amount of information may be acquired through the arrangement of the cameras.
However, a system that uses multiple cameras may entail a maintenance problem in proportion to the large amount of information that is acquired. In particular, the problem of camera calibration for detecting the position and posture of cameras increases the expense in proportion to the number of cameras.
Camera calibration includes a procedure for deriving intrinsic parameters and extrinsic parameters. Most computer vision systems that use cameras determine the place in which cameras are to be located and the direction that the cameras face, in the space designated by the designer of the corresponding computer vision system. In the case of multiple cameras, the positions and postures of all of the cameras are represented by the same coordinate system, and thus camera calibration is an essential process in image processing systems which use multiple cameras, such as in stereo imaging. In particular, it is expected that stereo cameras or multiple cameras are to be introduced in mobile phones or vehicles. In accordance with this expectation, manufacturing equipment for enabling camera manufacturers to calibrate multiple cameras is critically required.
As the number of calibration objects present in a calibration pattern increases, the accuracy of the results of a calibration algorithm is further improved. Most calibration programs automatically detect calibration objects, but require manual operations, such as the input of preliminary knowledge or the designation of a region of interest, as a precondition for the detection. Such manual operations need only to be performed once in the case of a single camera, but need to be performed several times in the case of multiple cameras, thus incurring a lot of expense.
In typical manufacturing processes, test equipment is limited as to the size and shape thereof. Test equipment is implemented in packages, for example, a package having a size and shape similar to a household refrigerator. The characteristics of the test equipment may improve the maintenance efficiency of the test equipment. In contrast, the vision system of stereo cameras is intended to identify objects located within a distance of about 10 m, with a baseline equal to the difference in position between the two eyes (binocular disparity). Here, the difference between the eyes may be about 6.5 cm. Therefore, the result of calibration by the vision system of stereo cameras may be trusted only when an image of a calibration object disposed at a representative location within a distance of about 10 m is captured and the result of the capturing is corrected.
In the manufacturing process, an apparatus for performing calibration of multiple cameras having a limited baseline must overcome the following restrictions:
1) The highest calibration quality must be derived via a smaller number of capturing operations.
2) To improve the ease of maintenance of the overall apparatus, lower components constituting the overall apparatus must be operated separately as far as possible.
3) The entire calibration process must be automated.